Association of largest meal timing and eating frequency with body mass index and obesity.

BACKGROUND AND AIMS: The circadian pattern of eating behaviors has garnered increasing interest as a strategy for obesity prevention and weight loss. It is believed that the benefits stem from aligning food intake with the body's natural daily rhythms. However, the existing body of evidence is limited in scale and scope and there has been insufficient evaluation of temporal eating behaviors, such as the specific time of day in which the highest calorie consumption occurs, meal frequency, and distribution. This research aims to explore the association between the timing of the largest meal of the day and eating frequency with Body Mass Index (BMI) and obesity. METHODS: Participants (n = 2050, 18-65y) were part of an exploratory cross-sectional and population-based research, with data collection in a virtual environment. Linear regression analyses and restricted cubic splines evaluated differences in BMI associated with independent eating variables [timing of the largest meal, number of meals/day (as continuous and categorical: ≤3 or >3/day), and each largest meal of the day (breakfast/lunch/dinner)]. Logistic regression models were fitted to assess Odds Ratios (OR) and 95 % Confidence Intervals (CI) of obesity associated with the same independent variables. RESULTS: Our main findings were that the timing of the largest meal and reporting dinner as the largest meal were associated with higher values of BMI (respectively, 0.07 kg/m2 and 0.85 kg/m2) and increased odds of obesity [respectively OR(95%CI):1.04(1.01,1.08), and OR(95%CI):1.67(1.18,2.38)]. Those who realized more than 3 meals/day presented lower values of BMI (-0.14 kg/m2) and 32 % lower odds of having obesity [OR(95%CI):0.68(0.52,0.89)]. Reporting lunch as the largest meal also protected against obesity [OR(95%CI):0.71(0.54,0.93)]. These associations were statistically significant and independent of sex, age, marital status, education level, diet quality, sleep duration, and weekly frequency of physical exercise. CONCLUSION: Having the largest meal earlier in the day, concentrating the majority of caloric intake during lunch, and consuming more than three meals a day, may present a promising intervention for preventing and treating obesity/overweight.

Association between sleep duration and latency, nocturnal awakenings, and body mass index among infants / Associação entre duração e latência do sono, despertares noturnos e índice de massa corporal de lactentes

ABSTRACT Objective: To investigate the association between sleep duration, nocturnal awakenings, and sleep latency with body mass index (BMI) at six and 12 months of age. Methods: 179 children from a birth cohort were enrolled. At six and 12 months of age, anthropometric data were obtained using standardized techniques and infants' mothers answered the Brief Infant Sleep Questionnaire for sleep data. The association of BMI with the independent variables (sleep duration, latency, and nocturnal awakenings) was assessed by linear regression models. Analyses were adjusted for potential confounders and a p-value<0.05 was adopted to define statistical significance. Results: For each additional hour of sleep duration, BMI was reduced by 0.15 kg/m² (95% confidence interval [CI] -0.28; -0.01; p=0.03) and each additional minute of sleep latency increased BMI by 0.01 kg/m² (95%CI -0.00; 0.03; p=0.02). These associations were independent of gestational age, child sex, birth weight, duration of exclusive breastfeeding, smoking during pregnancy, and mother's BMI, education, and marital status. Nocturnal awakenings showed no association with the outcome. Conclusions: Our findings suggest that sleep duration and sleep latency time are associated with BMI in the first year of life. Insights into the influence of sleep early in life on weight status may be helpful to complement future nutritional recommendations and prevent and treat obesity.

RESUMO Objetivo: Investigar a associação entre duração do sono, despertares noturnos e latência do sono com o índice de massa corporal (IMC) aos seis e 12 meses de idade. Métodos: foram incluídas 179 crianças de uma coorte de nascimentos. Aos seis e 12 meses de idade, dados antropométricos foram obtidos por meio de técnicas padronizadas e as mães dos lactentes responderam ao Brief Infant Sleep Questionnaire para dados do sono. A associação do IMC com as variáveis independentes (duração do sono, latência e despertares noturnos) foi avaliada por modelos de regressão linear. As análises foram ajustadas para potenciais fatores de confusão e o p-valor<0,05 foi adotado para definir a significância estatística. Resultados: Para cada hora adicional de duração do sono, o IMC foi reduzido em 0,15 kg/m² (intervalo de confiança [IC]95% -0,28; -0,01; p=0,03) e cada minuto adicional no tempo de latência resultou em aumento de 0,01 kg/m² (IC95% -0,00; 0,03; p=0,02) no IMC. Essas associações foram independentes da idade gestacional, sexo da criança, peso ao nascer, duração do aleitamento materno exclusivo, tabagismo durante a gravidez e IMC, escolaridade e estado civil da mãe. Os despertares noturnos não apresentaram associação com o desfecho. Conclusões: Nossos achados sugerem que a duração e a latência do sono estão associadas ao IMC no primeiro ano de vida. Informações sobre a influência do sono no início da vida sobre o status do peso podem ser úteis para complementar futuras recomendações nutricionais e prevenir e tratar a obesidade.

Association between sleep duration and latency, nocturnal awakenings, and body mass index among infants.

OBJECTIVE: To investigate the association between sleep duration, nocturnal awakenings, and sleep latency with body mass index (BMI) at six and 12 months of age. METHODS: 179 children from a birth cohort were enrolled. At six and 12 months of age, anthropometric data were obtained using standardized techniques and infants' mothers answered the Brief Infant Sleep Questionnaire for sleep data. The association of BMI with the independent variables (sleep duration, latency, and nocturnal awakenings) was assessed by linear regression models. Analyses were adjusted for potential confounders and a p-value<0.05 was adopted to define statistical significance. RESULTS: For each additional hour of sleep duration, BMI was reduced by 0.15 kg/m² (95% confidence interval [CI] -0.28; -0.01; p=0.03) and each additional minute of sleep latency increased BMI by 0.01 kg/m² (95%CI -0.00; 0.03; p=0.02). These associations were independent of gestational age, child sex, birth weight, duration of exclusive breastfeeding, smoking during pregnancy, and mother's BMI, education, and marital status. Nocturnal awakenings showed no association with the outcome. CONCLUSIONS: Our findings suggest that sleep duration and sleep latency time are associated with BMI in the first year of life. Insights into the influence of sleep early in life on weight status may be helpful to complement future nutritional recommendations and prevent and treat obesity.

Association between iron deficiency anemia and sleep duration in the first year of life.

OBJECTIVE: The aim of this study was to investigate the association between iron deficiency anemia and sleep duration in the first year of life. METHODS: A total of 123 infants were investigated, with sleep being evaluated at 3, 6, and 12 months of age and anemia at birth and 6 months. The cutoff points for anemia and short sleep duration were hemoglobin <11 g/dL (at birth and/or 6 months) and <10 h (at 3, 6, and 12 months), respectively. The comparison of the average sleep time between infants with and without anemia was performed using the Student's t-test, and logistic regression models were also used to verify differences in the sleep duration (short/not short) between the groups. Linear regression analyses were conducted to determine the association between sleep duration and hemoglobin values. The analyses were adjusted for potential confounders. RESULTS: Children with anemia were more likely to be short sleepers [odds ratio (95% confidence interval (CI)): 4.02 (1.02-15.76); p≤0.05], and for each unit increase in hemoglobin values, the sleep duration increased by 16.2 min [ß (95%CI): 0.27 (0.00-0.55); p≤0.05), regardless of family income, maternal schooling, gender, and body mass index at birth. CONCLUSIONS: Our results suggest that iron deficiency anemia is associated with short sleep duration in the first year of life and indicate the need for longitudinal investigations, with longer follow-up, to verify the impact of anemia on sleep duration at subsequent ages.

Association between iron deficiency anemia and sleep duration in the first year of life / Associação entre anemia ferropriva e duração do sono no primeiro ano de vida

Abstract Objective: The aim of this study was to investigate the association between iron deficiency anemia and sleep duration in the first year of life. Methods: A total of 123 infants were investigated, with sleep being evaluated at 3, 6, and 12 months of age and anemia at birth and 6 months. The cutoff points for anemia and short sleep duration were hemoglobin <11 g/dL (at birth and/or 6 months) and <10 h (at 3, 6, and 12 months), respectively. The comparison of the average sleep time between infants with and without anemia was performed using the Student's t-test, and logistic regression models were also used to verify differences in the sleep duration (short/not short) between the groups. Linear regression analyses were conducted to determine the association between sleep duration and hemoglobin values. The analyses were adjusted for potential confounders. Results: Children with anemia were more likely to be short sleepers [odds ratio (95% confidence interval (CI)): 4.02 (1.02-15.76); p≤0.05], and for each unit increase in hemoglobin values, the sleep duration increased by 16.2 min [β (95%CI): 0.27 (0.00-0.55); p≤0.05), regardless of family income, maternal schooling, gender, and body mass index at birth. Conclusions: Our results suggest that iron deficiency anemia is associated with short sleep duration in the first year of life and indicate the need for longitudinal investigations, with longer follow-up, to verify the impact of anemia on sleep duration at subsequent ages.

RESUMO Objetivo: Investigar a associação entre a anemia por deficiência de ferro e a duração do sono no primeiro ano de vida. Métodos: Foram avaliadas 123 crianças, sendo o sono investigado aos três, seis e 12 meses de idade e a anemia ao nascimento e aos seis meses. Utilizaram-se como pontos de corte para anemia e curta duração de sono, respectivamente, hemoglobina<11 g/dL (nascimento e/ou seis meses) e tempo total <10 h (3, 6 e/ou 12 meses). A comparação do tempo médio de sono entre as crianças com e sem anemia foi realizada pelo teste t de Student e modelos de regressão logística foram usados para verificar diferenças na duração do sono (curta/não curta) entre os grupos. Análises de regressão linear foram conduzidas para determinar a associação entre a duração do sono e valores de hemoglobina. As análises foram ajustadas para potenciais confundidores. Resultados: As crianças com anemia tiveram maior chance de apresentar curta duração do sono [odds ratio — OR (intervalo de confiança — IC95%): 4,02 (1,02-15,76); p≤0,05]. Para cada unidade de aumento nos valores da hemoglobina, o tempo de sono aumentou em 16,2 min [β (IC95%): 0,27 (0,00-0,55); p≤0,05), independentemente de renda familiar, escolaridade materna, sexo e índice de massa corporal ao nascimento. Conclusões: Nossos resultados sugerem que a anemia ferropriva está associada à curta duração do sono no primeiro ano de vida e indicam a necessidade de investigações longitudinais, com maior tempo de seguimento, para verificar o impacto da anemia na duração do sono em idades subsequentes.

Consumo de alimentos açucarados e qualidade do sono no primeiro ano de vida: dados de uma coorte de nascimentos no nordeste do Brasil / Consumo de alimentos azucarados y calidad del sueño en el primer año de vida: datos de una cohorte de nacimiento en el noreste de Brasil / Consumption of sugary foods and quality of sleep in the first year of life: data from a birth cohort in northeastern Brazil

RESUMO Objetivo: investigar a duração do sono (DS), a frequência de despertares noturnos (DN) e o consumo de alimentos açucarados no primeiro ano de vida e verificar a associação entre o consumo desses alimentos e a má qualidade do sono. Métodos: a população do estudo foi composta de 179 crianças integrantes de uma coorte de nascimentos de Rio Largo-AL. As mães foram questionadas sobre a oferta regular de alimentos açucarados (açúcar/farinhas de cereais instantâneas com açúcar/bebidas açucaradas/doces) e o sono foi investigado pelo questionário traduzido e validado Brief Infant Sleep Questionnaire. Foram considerados indicadores de má qualidade do sono DS<12h e DN>2. Os testes de qui-quadrado de Pearson e exato de Fisher foram adotados para verificar associações entre o consumo de açucarados e a má qualidade do sono aos seis e 12 meses (p<0,05). Resultados: mais da metade das crianças apresentou DS<1 2h (60,3%) e cerca de » DN>2. O consumo regular de pelo menos uma das categorias de açucarados foi verificado entre 50,6, 91,1 e 100% das crianças aos três, seis e 12 meses de idade, respectivamente. Não foram encontradas associações entre o consumo desses alimentos e os indicadores de má qualidade de sono. Conclusão: o consumo de açucarados e a má qualidade de sono foram frequentes em nosso estudo, no entanto, não se identificou associação entre as variáveis. Mais investigações são necessárias para elucidar como o sono e a alimentação se inter-relacionam e se potencializam mutuamente como fatores determinantes do crescimento e desenvolvimento de lactentes.

RESUMEN Objetivo: investigar la duración del sueño (DS), la frecuencia de despertares nocturnos (DN) y el consumo de alimentos azucarados en el primer año de vida y verificar la asociación entre el consumo de estos alimentos y la mala calidad del sueño. Métodos: la población de estudio consistió en 179 niños de una cohorte de nacimiento en Rio Largo-AL. Se preguntó a las madres sobre el suministro regular de alimentos azucarados (azúcar / harinas de cereales instantáneas con azúcar / bebidas endulzadas / dulces) y se investigó el sueño mediante el cuestionario traducido y validado BriefInfantSleepQuestionnaire. Fueron considerados indicadores de mala calidad del sueño DS <12h y DN> 2. Se utilizaron las pruebas de chi-cuadrado de Pearson y exacta de Fisher para verificar las asociaciones entre el consumo de azúcar y la mala calidad del sueño a los seis y 12 meses (p <0.05). Resultados: más de la mitad de los niños tenían DS <1 2 h (60,3%) y alrededor de » DN> 2. Se verificó el consumo regular de al menos una de las categorías azucaradas entre el 50,6, el 91,1 y el 100% de los niños a los tres, seis y 12 meses de edad, respectivamente. No se encontraron asociaciones entre el consumo de estos alimentos y los indicadores de mala calidad del sueño. Conclusión: el consumo de azucarados y la mala calidad del sueño fueron frecuentes en nuestro estudio, sin embargo, no se identificó asociación entre las variables. Se necesita más investigación para dilucidar cómo el sueño y la alimentación se interrelacionan y se mejoran mutuamente como determinantes del crecimiento y desarrollo infantil.

ABSTRACT Objective: to investigate sleep duration (SD), frequency of night awakenings (NA) and consumption of sugary foods in the first year of life and to verify the association between consumption of these foods and poor sleep quality. Methods: the study population consisted of 179 children from a birth cohort in Rio Largo-AL. Mothers were asked about the regular supply of sugary foods (sugar/instant cereal flours with sugar/sweetened drinks/sweets) and sleep was investigated by the translated and validated Brief Infant Sleep Questionnaire. DS<12h and AN>2 were considered indicators of poor sleep quality. Pearson's chi-square and Fisher's exact tests were used to verify associations between sugary consumption and poor sleep quality at six and 12 months (p<0.05). Results: more than half of the children had SD<1 2h (60.3%) and about » AN>2. Regular consumption of at least one of the sugary categories was verified among 50.6, 91.1 and 100% of children at three, six and 12 months of age, respectively. No associations were found between the consumption of these foods and indicators of poor sleep quality. Conclusion: sugary consumption and poor sleep quality were frequent in our study; however, no association was identified between the variables. More investigations are needed to elucidate how sleep and feeding are interrelated and mutually potentiate as determinants of infant growth and development.

[Time-series analysis of the consumption of sweetened soft drinks among adults in Brazil: 2007 to 2014]. / Análise de série temporal do consumo de bebidas açucaradas entre adultos no Brasil: 2007 a 2014.

The scope of this article is to analyze the time-series trend and factors associated with the consumption of soft drinks or packaged fruit juices among adults in Brazil. It is a study based on secondary data from the System of Surveillance of Risk Factors and Protection for Chronic Diseases by Telephone Survey conducted among Brazilian adults between 2007 and 2014. The consumption frequency and intensity (number of cups or cans per week) of soda or packaged juice was checked. Socio-demographic and behavioral data were the independent variables. The time-series trend of annual consumption was evaluated by means of Linear Regression. The factors (age, sex, region, work, schooling and TV screen time) associated with the consumption of these beverages were investigated by Poisson regression. There was a 32.7% reduction in soft drink or packaged juice consumption between 2007 and 2014. Factors associated with higher consumption were: male sex (p = 0.000); 18-29 year-age-range (p = 0.000); residence in the central-west, southeast and southern regions (p = 0.000); lower schooling (p = 0.616); being employed (p = 0.007) and more than 3 hours of TV screen time per day (p = 0.000). The analyses describe a downward trend in the consumption of soda or packaged fruit juice among adults in Brazil from 2007 to 2014.

O objetivo deste artigo é analisar a tendência temporal e os fatores associados ao consumo de refrigerante ou suco artificial entre adultos no Brasil. Estudo desenvolvido a partir de dados secundários do Sistema de Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico, realizado com adultos brasileiros entre 2007-2014. Foi verificada a frequência e a intensidade do consumo (quantidade de copos ou latas por semana) de refrigerante ou suco artificial. Dados sociodemográficos e comportamentais foram as variáveis independentes. A tendência temporal do consumo anual foi avaliada por meio de Regressão Linear. Os fatores associados (idade, sexo, região, trabalho, escolaridade, hábito de assistir TV) ao consumo dessas bebidas foram investigados por Regressão de Poisson. Houve redução de 32,7% do consumo de refrigerante ou suco artificial entre 2007 e 2014. Os fatores associados ao maior consumo foram: sexo masculino (p = 0,000); faixa etária de 18-29 anos (p = 0,000); residência nas regiões centro-oeste, sudeste e sul (p = 0,000); menor escolaridade (p = 0,616); estar empregado (p = 0,007) e assistir TV mais de 3 horas por dia (p = 0,000). As análises descrevem uma tendência de queda no consumo de refrigerante ou suco artificial entre os adultos no Brasil de 2007 a 2014.

Análise de série temporal do consumo de bebidas açucaradas entre adultos no Brasil: 2007 a 2014 / Time-series analysis of the consumption of sweetened soft drinks among adults in Brazil: 2007 to 2014

Resumo O objetivo deste artigo é analisar a tendência temporal e os fatores associados ao consumo de refrigerante ou suco artificial entre adultos no Brasil. Estudo desenvolvido a partir de dados secundários do Sistema de Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico, realizado com adultos brasileiros entre 2007-2014. Foi verificada a frequência e a intensidade do consumo (quantidade de copos ou latas por semana) de refrigerante ou suco artificial. Dados sociodemográficos e comportamentais foram as variáveis independentes. A tendência temporal do consumo anual foi avaliada por meio de Regressão Linear. Os fatores associados (idade, sexo, região, trabalho, escolaridade, hábito de assistir TV) ao consumo dessas bebidas foram investigados por Regressão de Poisson. Houve redução de 32,7% do consumo de refrigerante ou suco artificial entre 2007 e 2014. Os fatores associados ao maior consumo foram: sexo masculino (p = 0,000); faixa etária de 18-29 anos (p = 0,000); residência nas regiões centro-oeste, sudeste e sul (p = 0,000); menor escolaridade (p = 0,616); estar empregado (p = 0,007) e assistir TV mais de 3 horas por dia (p = 0,000). As análises descrevem uma tendência de queda no consumo de refrigerante ou suco artificial entre os adultos no Brasil de 2007 a 2014.

Abstract The scope of this article is to analyze the time-series trend and factors associated with the consumption of soft drinks or packaged fruit juices among adults in Brazil. It is a study based on secondary data from the System of Surveillance of Risk Factors and Protection for Chronic Diseases by Telephone Survey conducted among Brazilian adults between 2007 and 2014. The consumption frequency and intensity (number of cups or cans per week) of soda or packaged juice was checked. Socio-demographic and behavioral data were the independent variables. The time-series trend of annual consumption was evaluated by means of Linear Regression. The factors (age, sex, region, work, schooling and TV screen time) associated with the consumption of these beverages were investigated by Poisson regression. There was a 32.7% reduction in soft drink or packaged juice consumption between 2007 and 2014. Factors associated with higher consumption were: male sex (p = 0.000); 18-29 year-age-range (p = 0.000); residence in the central-west, southeast and southern regions (p = 0.000); lower schooling (p = 0.616); being employed (p = 0.007) and more than 3 hours of TV screen time per day (p = 0.000). The analyses describe a downward trend in the consumption of soda or packaged fruit juice among adults in Brazil from 2007 to 2014.

Weight gain and changes in nutritional status of Brazilian adults after 20 years of age: a time-trend analysis (2006-2012). / Ganho de peso e mudança do estado nutricional de brasileiros após os 20 anos de idade: uma análise de série temporal (2006-2012).

OBJECTIVE: To analyze weight gain (WG) and change in nutritional status (NS) after the age of 20 years in the Brazilian adult population between 2006 and 2012. METHODS: Time series using seven surveys from the Surveillance of Risk and Protective Factors for Chronic Diseases by Telephone Survey (VIGITEL). The analyses were stratified by region, age, sex and education, considering the sampling weights and complex design. In addition, relative weight change (RWC) in the period was determined for each category of independent variables using linear regression models. RESULTS: Analyses showed an increase in WG after 20 years in two vectors: by survey year and age group, where the increase was higher in women. From 2006 to 2012, individuals 25-29 years old (women: RWC = 70%; ßyear = 0.54 kg/year) and 30-34 years old (women: RWC = 56%; ßyear = 0.57 kg/year) showed greater RWC. In 2012, the higher WG occurred in the age groups of 21-24 and 25-29 years old. Regarding the change in NS, individuals who were overweight at 20 years had a higher probability of remaining in this condition (or shifting to obesity) over time. However, among those who had a normal weight, the probability of not becoming overweight or obese was > 80%, independently of sex. CONCLUSION: The Brazilian population displayed progressive WG in adulthood, especially in the first decade after the age of 20, in addition to the period effect. On the other hand, individuals with normal weight in their 20s tended to maintain the same condition.

PRE- AND PERINATAL FACTORS ASSOCIATED WITH WEIGHT GAIN AMONG PRESCHOOL CHILDREN ENROLLED AT DAY CARE CENTERS.

OBJECTIVE: To identify the factors associated with excessive weight gain in preschool children enrolled at daycare centers in a capital of the Northeast region of Brazil. METHODS: It was a cross-sectional study conducted at the five daycare centers located in the city's district of most socioeconomic vulnerability. The study included 326 preschool children (17 to 63 months old) from both genders. The dependent variable was the conditional weight gain (CWG), that represents how much a child, according to their gender, deviated from their peers in relation to the expected weight gain, given sample's birthweight, gender, and age at the survey. Univariate tests (t-test and analysis of variance) were used to compare CWG means according to environmental and biological factors, considering the independent variables with p<0.20 as electable for the multiple linear regression model. In the final model, variables with p<0.05 or that contributed to the model adjustment were kept. RESULTS: Children's mean age was 45.4±9.9 months, and 53.4% of the sample consisted of boys. The prevalence of overweight was 7%. In the multivariable linear regression model, it was possible to identify that the following factors were associated with excessive weight gain among preschool children: less than six prenatal care visits (0.36 SD [95%CI 0.13-0.60]), not rooming-in in the postpartum period (0.30 SD [95%CI 0.03-0.58]), and never breastfed (0.44 SD [95%CI 0.06-0.81]). CONCLUSIONS: Inadequate prenatal (appointments) and perinatal care (mother-infant rooming-in and absence of breastfeeding) were associated with excessive weight gain among low-income preschool children.

Pre- and perinatal factors associated with weight gain among preschool children enrolled at day care centers / Associação entre fatores pré e perinatais e padrão de ganho de peso em pré-escolares de centros de educação infantil

ABSTRACT Objective: To identify the factors associated with excessive weight gain in preschool children enrolled at daycare centers in a capital of the Northeast region of Brazil. Methods: It was a cross-sectional study conducted at the five daycare centers located in the city's district of most socioeconomic vulnerability. The study included 326 preschool children (17 to 63 months old) from both genders. The dependent variable was the conditional weight gain (CWG), that represents how much a child, according to their gender, deviated from their peers in relation to the expected weight gain, given sample's birthweight, gender, and age at the survey. Univariate tests (t-test and analysis of variance) were used to compare CWG means according to environmental and biological factors, considering the independent variables with p<0.20 as electable for the multiple linear regression model. In the final model, variables with p<0.05 or that contributed to the model adjustment were kept. Results: Children's mean age was 45.4±9.9 months, and 53.4% of the sample consisted of boys. The prevalence of overweight was 7%. In the multivariable linear regression model, it was possible to identify that the following factors were associated with excessive weight gain among preschool children: less than six prenatal care visits (0.36 SD [95%CI 0.13-0.60]), not rooming-in in the postpartum period (0.30 SD [95%CI 0.03-0.58]), and never breastfed (0.44 SD [95%CI 0.06-0.81]). Conclusions: Inadequate prenatal (appointments) and perinatal care (mother-infant rooming-in and absence of breastfeeding) were associated with excessive weight gain among low-income preschool children.

RESUMO Objetivo: Identificar fatores associados com o ganho ponderal excessivo entre pré-escolares de Centros de Educação Infantil (CEIs) em uma capital do Nordeste brasileiro. Métodos: Estudo transversal realizado em cinco CEIs situados no distrito de maior vulnerabilidade socioeconômica do município. Foram incluídas 326 crianças de ambos os sexos, com idades entre 17 e 63 meses. A variável dependente foi a evolução ponderal condicional (EPC), a qual representa o quanto uma criança desviou do ganho de peso esperado em relação a seus pares, de acordo com o sexo, o peso ao nascer e a idade no inquérito. Análises univariadas (teste t de Student ou análise de variância) foram utilizadas para comparar as médias da EPC em função de fatores biológicos e ambientais, considerando como elegíveis para o modelo múltiplo de regressão linear as variáveis independentes com p<0,20; permaneceram no modelo final aquelas que apresentaram p<0,05 ou que contribuíram no ajuste do modelo. Resultados: A média de idade foi de 45,4±9,9 meses e 53,4% eram meninos. A prevalência de excesso de peso foi de 7%. No modelo múltiplo de regressão linear, identificamos que realizar menos de 6 consultas pré-natal (DP=0,36; IC95% 0,13-0,60), não ter permanecido em alojamento conjunto no pós-parto (DP=0,30; IC95% 0,03-0,58) e nunca ter sido amamentado (DP=0,44; IC95% 0,06-0,81) foram fatores associados com o ganho excessivo de peso entre pré-escolares. Conclusões: A inadequação dos cuidados pré-natal (consultas) e perinatal (não permanecer em alojamento conjunto e ausência da amamentação) se associaram com o ganho excessivo de peso entre pré-escolares de baixa renda.

Ganho de peso e mudança do estado nutricional de brasileiros após os 20 anos de idade: uma análise de série temporal (2006-2012) / Weight gain and changes in nutritional status of Brazilian adults after 20 years of age: a time-trend analysis (2006-2012)

RESUMO: Objetivos: Analisar o ganho de peso (GP) e a mudança do estado nutricional (EN) após os 20 anos de idade na população brasileira entre os anos de 2006 e 2012. Metodologia: Série temporal com base em sete inquéritos transversais do Sistema de Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico (VIGITEL). As análises foram estratificadas por região, faixa etária, sexo e escolaridade, considerando-se o plano e a ponderação amostral. Ainda, estimou-se a variação ponderal relativa (VPR) no período para os estratos analisados por meio de modelos de regressão linear. Resultados: Identificaram-se dois vetores de aumento no GP após os 20 anos de idade: ano do inquérito e faixa etária, que foram mais expressivos entre as mulheres. Entre 2006 e 2012, as faixas etárias que apresentaram a maior variação temporal foram de 25-29 (mulheres: VPR = 70%; βano = 0,54 kg/ano) e 30-34 anos (mulheres: VPR = 56%; βano = 0,57 kg/ano). Em 2012, o maior aumento de GP ocorreu nas faixas etárias de 21-24 e 25-29 anos. Quanto à mudança de EN, indivíduos que apresentavam excesso de peso aos 20 anos tiveram maior probabilidade de permanecer nessa condição (ou migrar do sobrepeso para obesidade) com o avançar da idade. Contudo, entre os que eram eutróficos, a probabilidade de permanecer com o mesmo EN foi > 80%, independentemente do sexo. Conclusão: Além do efeito do tempo, a população brasileira apresentou progressivo GP no decorrer da fase adulta, sobretudo na primeira década após os 20 anos. Por outro lado, indivíduos eutróficos aos 20 anos tenderam a permanecer nessa condição.

ABSTRACT: Objective: To analyze weight gain (WG) and change in nutritional status (NS) after the age of 20 years in the Brazilian adult population between 2006 and 2012. Methods: Time series using seven surveys from the Surveillance of Risk and Protective Factors for Chronic Diseases by Telephone Survey (VIGITEL). The analyses were stratified by region, age, sex and education, considering the sampling weights and complex design. In addition, relative weight change (RWC) in the period was determined for each category of independent variables using linear regression models. Results: Analyses showed an increase in WG after 20 years in two vectors: by survey year and age group, where the increase was higher in women. From 2006 to 2012, individuals 25-29 years old (women: RWC = 70%; βyear = 0.54 kg/year) and 30-34 years old (women: RWC = 56%; βyear = 0.57 kg/year) showed greater RWC. In 2012, the higher WG occurred in the age groups of 21-24 and 25-29 years old. Regarding the change in NS, individuals who were overweight at 20 years had a higher probability of remaining in this condition (or shifting to obesity) over time. However, among those who had a normal weight, the probability of not becoming overweight or obese was > 80%, independently of sex. Conclusion: The Brazilian population displayed progressive WG in adulthood, especially in the first decade after the age of 20, in addition to the period effect. On the other hand, individuals with normal weight in their 20s tended to maintain the same condition.

[Temporal trend and factors associated with consumption of fatty meats by the Brazilian population between 2007 and 2014]. / Tendência temporal e fatores associados ao consumo de carnes gordurosas na população brasileira entre de 2007 a 2014.

The scope of this article is to analyze the temporal trend of the consumption of fatty meats and factors associated with the consumption thereof between 2007 and 2014. A time series of fatty meat consumption (red meat and chicken) was conducted by the "Surveillance of Risk and Protection Factors for Chronic Diseases (Vigitel) telephone survey." The consumption trend was analyzed by joinpoint regression and expressed in annual percentage variation (VPA [95% CI]), while sociodemographic and behavioral factors were associated in 2007 and 2014 by the generalized linear (log-binomial) model. Between 2007 and 2014, white meat consumption ≥ 5x/ week increased (VPA 6.9% [5.7, 8.1]), while red meat consumption remained stable. Fatty meat consumption showed a significant reduction as of 2011 (VPA -4% [-7.5; -0.2]), probably due to the reduction in the consumption of fatty red meats. It was found that males, of younger age groups living in the center-west/south-east/south regions, with low education level and alcohol and tobacco abuse were associated with fatty meat consumption; while watching television>3hours/day was only associated in 2014. The consumption of fatty meats showed a reduction trend in the Brazilian population, and its consumption was associated with sociodemographic and behavioral variables.

O objetivo deste artigo é analisar a tendência temporal do consumo de carnes gordurosas (CG) e fatores associados ao seu consumo habitual entre 2007 e 2014. Série temporal do consumo de CG (vermelha e frango/galinha) realizada a partir do inquérito telefônico "Vigilância de Fatores de Risco e Proteção para Doenças Crônicas" (Vigitel). A tendência do consumo foi analisada por regressão "joinpoint" e expressa em variação percentual anual (VPA[IC95%]), enquanto os fatores sociodemográficos e comportamentais associados em 2007 e 2014 por modelo linear generalizado (log-binomial). Entre 2007-2014, o consumo ≥ 5x/semana de carnes brancas aumentou (VPA 6,9% [5,7; 8,1]), enquanto as vermelhas permaneceu estável. O consumo CG apresentou redução significativa a partir de 2011 (VPA -4,0%[-7,5; -0,2]), provavelmente em função da redução no consumo das vermelhas gordurosas. Identificamos que o sexo masculino, as faixas etárias mais jovens, residir nas regiões CO/SE/S, escolaridade ≤ 8 anos e o uso abusivo de álcool e tabaco se associaram com o consumo de CG; assistir televisão > 3h/dia foi associada apenas em 2014. No período analisado houve redução no consumo de CG pela população brasileira, sendo seu consumo associado a fatores sociodemográficas e outros comportamentos promotores de DCNT.

Tendência temporal e fatores associados ao consumo de carnes gordurosas na população brasileira entre de 2007 a 2014 / Temporal trend and factors associated with consumption of fatty meats by the Brazilian population between 2007 and 2014

Resumo O objetivo deste artigo é analisar a tendência temporal do consumo de carnes gordurosas (CG) e fatores associados ao seu consumo habitual entre 2007 e 2014. Série temporal do consumo de CG (vermelha e frango/galinha) realizada a partir do inquérito telefônico "Vigilância de Fatores de Risco e Proteção para Doenças Crônicas" (Vigitel). A tendência do consumo foi analisada por regressão "joinpoint" e expressa em variação percentual anual (VPA[IC95%]), enquanto os fatores sociodemográficos e comportamentais associados em 2007 e 2014 por modelo linear generalizado (log-binomial). Entre 2007-2014, o consumo ≥ 5x/semana de carnes brancas aumentou (VPA 6,9% [5,7; 8,1]), enquanto as vermelhas permaneceu estável. O consumo CG apresentou redução significativa a partir de 2011 (VPA -4,0%[-7,5; -0,2]), provavelmente em função da redução no consumo das vermelhas gordurosas. Identificamos que o sexo masculino, as faixas etárias mais jovens, residir nas regiões CO/SE/S, escolaridade ≤ 8 anos e o uso abusivo de álcool e tabaco se associaram com o consumo de CG; assistir televisão > 3h/dia foi associada apenas em 2014. No período analisado houve redução no consumo de CG pela população brasileira, sendo seu consumo associado a fatores sociodemográficas e outros comportamentos promotores de DCNT.

Abstract The scope of this article is to analyze the temporal trend of the consumption of fatty meats and factors associated with the consumption thereof between 2007 and 2014. A time series of fatty meat consumption (red meat and chicken) was conducted by the "Surveillance of Risk and Protection Factors for Chronic Diseases (Vigitel) telephone survey." The consumption trend was analyzed by joinpoint regression and expressed in annual percentage variation (VPA [95% CI]), while sociodemographic and behavioral factors were associated in 2007 and 2014 by the generalized linear (log-binomial) model. Between 2007 and 2014, white meat consumption ≥ 5x/ week increased (VPA 6.9% [5.7, 8.1]), while red meat consumption remained stable. Fatty meat consumption showed a significant reduction as of 2011 (VPA -4% [-7.5; -0.2]), probably due to the reduction in the consumption of fatty red meats. It was found that males, of younger age groups living in the center-west/south-east/south regions, with low education level and alcohol and tobacco abuse were associated with fatty meat consumption; while watching television>3hours/day was only associated in 2014. The consumption of fatty meats showed a reduction trend in the Brazilian population, and its consumption was associated with sociodemographic and behavioral variables.

Anemia and associated factors in women at reproductive age in a Brazilian Northeastern municipality. / Anemia e fatores associados em mulheres de idade reprodutiva de um município do Nordeste brasileiro.

OBJECTIVE: To identify the prevalence of iron-deficiency anemia and associated factors in women at reproductive age in the city of Vitoria de Santo Antão, Pernambuco. METHODS: Cross-sectional study with a representative sample of 322 women aged between 15 and 49 years. Anemia was diagnosed by measuring hemoglobin below 12 g/dL. The independent variables were: demographic and socioeconomic conditions (age, race/color, education, literacy, presence of children aged < 5 years in the household, regime of residence occupation, ownership of assets, and per capita income), housing conditions (number of people in the household, number of bedrooms, and basic sanitation), food and nutrition security and women's health conditions (menarche, number of pregnancies, history of miscarriages, body mass index, and ferritin). Poisson regression was used to analyze the factors associated with anemia. RESULTS: The mean age of women surveyed was 31.1 ± 6.61 years old (standard deviation - SD) and the prevalence of anemia was 18.6% (95%CI 14.7 - 23.3). After multivariate analysis, anemia was significantly associated with age ≤ 19 years and low ferritin. CONCLUSION: The prevalence of anemia is characterized as a mild public health problem and follows the world trend in developing countries.

Prevalence and Factors Associated with Metabolic Syndrome among Brazilian Adult Population: National Health Survey - 2013.

BACKGROUND: In Brazil, population-based researches analyzing prevalence and factors associated with metabolic syndrome (MS), a recognized predictor of cardiovascular diseases (CVD), and an important cause of disability and death in the country are scarce. OBJECTIVE: To evaluate prevalence of MS and its associated factors in Brazilian population. METHODS: Secondary analysis of the 2013 National Health Survey, a cross-sectional survey with national representativeness of Brazilian adult population (n = 59,402). MS was the outcome variable, defined from harmonization of cardiology international consensus as load ≥ 3 of the following components: self-reported diabetes and hypercholesterolemia, high blood pressure and high waist circumference. Analysis were stratified by sex and prevalence ratios, with their respective 99% confidence intervals (PR [CI 99%]) calculated by simple and multiple Poisson regression models. RESULTS: MS prevalence was 8.9%, being significantly higher among women compared to men; in general, this pattern was maintained in relation to exposure variables studied. Additionally, less than 25% of population did not present any MS component. In final multiple models, sociodemographic, behavioral and comorbidity variables were associated with MS, however, while low schooling (1.46 [1.23-1.74], cerebrovascular accident (1.36 [1], 00] (1.28 [1.03-1.62]) were associated among women, chronic renal failure (1.85 [2.23-2.76]) was associated exclusively among men. CONCLUSION: We identified MS high prevalence in Brazilian population; on the other hand, factors associated with this condition were different depending on sex.

Prevalence and Factors Associated with Metabolic Syndrome among Brazilian Adult Population: National Health Survey - 2013 / Prevalência e Fatores Associados com a Síndrome Metabólica na População Adulta Brasileira: Pesquisa Nacional de Saúde - 2013

Abstract Background: In Brazil, population-based researches analyzing prevalence and factors associated with metabolic syndrome (MS), a recognized predictor of cardiovascular diseases (CVD), and an important cause of disability and death in the country are scarce. Objective: To evaluate prevalence of MS and its associated factors in Brazilian population. Methods: Secondary analysis of the 2013 National Health Survey, a cross-sectional survey with national representativeness of Brazilian adult population (n = 59,402). MS was the outcome variable, defined from harmonization of cardiology international consensus as load ≥ 3 of the following components: self-reported diabetes and hypercholesterolemia, high blood pressure and high waist circumference. Analysis were stratified by sex and prevalence ratios, with their respective 99% confidence intervals (PR [CI 99%]) calculated by simple and multiple Poisson regression models. Results: MS prevalence was 8.9%, being significantly higher among women compared to men; in general, this pattern was maintained in relation to exposure variables studied. Additionally, less than 25% of population did not present any MS component. In final multiple models, sociodemographic, behavioral and comorbidity variables were associated with MS, however, while low schooling (1.46 [1.23-1.74], cerebrovascular accident (1.36 [1], 00] (1.28 [1.03-1.62]) were associated among women, chronic renal failure (1.85 [2.23-2.76]) was associated exclusively among men. Conclusion: We identified MS high prevalence in Brazilian population; on the other hand, factors associated with this condition were different depending on sex.

Resumo Fundamento: No Brasil, são escassas as investigações de base populacional analisando a prevalência e os fatores associados à síndrome metabólica (SM), reconhecida preditora de doenças cardiovasculares (DCV), importante causa de invalidez e morte no país. Objetivo: Avaliar a prevalência da SM e seus fatores associados na população brasileira. Métodos: Análise secundária da Pesquisa Nacional de Saúde de 2013, inquérito transversal com representatividade nacional da população adulta brasileira (n = 59.402). A variável de desfecho foi a SM, definida a partir da harmonização dos consensos internacionais de cardiologia como carga ≥ 3 dos seguintes componentes: diabetes e hipercolesterolemia autorreferidas, pressão arterial elevada e circunferência da cintura elevada. As análises foram estratificadas por sexo e as razões de prevalência, com seus respectivos intervalos de confiança de 99% (RP[IC99%]), foram calculadas por meio de modelos regressão simples e múltiplos de Poisson. Resultados: A prevalência de SM foi 8,9%, sendo significativamente maior entre as mulheres em comparação aos homens; de modo geral, tal padrão manteve-se em relação as variáveis de exposição estudadas. Adicionalmente, menos de 25% da população não apresentou qualquer componente de SM. Nos modelos múltiplos finais, variáveis sociodemográficas, comportamentais e de comorbidades se associaram à SM, porém, enquanto a baixa escolaridade (1,46[1,23-1,74]), acidente vascular cerebral (1,36[1,00-1,86]) e outras DCV (1,29[1,03-1,62]) se associaram entre as mulheres, a insuficiência renal crônica (1,85[2,23-2,76]) se associou exclusivamente entre os homens. Conclusão: Identificamos elevada prevalência de SM na população brasileira, sendo que os fatores associados a essa condição são distintos em função do sexo.

[Trends and factors associated with food insecurity in Brazil: the National Household Sample Survey, 2004, 2009, and 2013]. / Tendência e fatores associados à insegurança alimentar no Brasil: Pesquisa Nacional por Amostra de Domicílios 2004, 2009 e 2013.

The aim of this study was to analyze trends and factors associated with food insecurity in Brazil in 2004, 2009, and 2013, using microdata from the National Household Sample Survey (PNAD). Food insecurity was assessed using the Brazilian Food Insecurity Scale. Independent variables were selected from a conceptual model of determination of food insecurity, which was also used in the elaboration of multiple generalized linear models. The results show a downward trend in food insecurity prevalence from 2004 to 2013, especially for moderate and severe food insecurity, from 17% in 2004 (95%CI: 15.7-18.4) to 7.9% in 2013 (95%CI: 7.2-8.7). Despite important decreases in the prevalence of moderate and severe food insecurity, regardless of the level of determination, the population strata with the lowest prevalence in 2004 showed the largest relative reduction. As for factors associated with moderate and severe food insecurity, they remained the same in the ten years covered by the PNAD survey, namely: the North and Northeast regions, urban areas with inadequate sanitation, household density > 2 persons per bedroom, ≤ 4 household durable consumer goods, and households headed by females, individuals < 60 years, and non-whites, ≤ 4 years of schooling, and being unemployed. From 2004 to 2013, the prevalence of Brazilian households with moderate and severe food insecurity dropped by half, but from the perspective of equity the advances occurred unequally and were lower in strata with greater social, economic, and demographic vulnerability.

O objetivo deste estudo foi analisar a tendência e fatores associados à insegurança alimentar no Brasil nos anos de 2004, 2009 e 2013, utilizando microdados da Pesquisa Nacional por Amostra de Domicílios (PNAD). A insegurança alimentar foi avaliada por meio da Escala Brasileira de Insegurança Alimentar. As variáveis independentes foram selecionadas a partir de modelo conceitual de determinação da insegurança alimentar, sendo esse também utilizado para a elaboração dos modelos lineares generalizados múltiplos. Os resultados descrevem tendência de redução na prevalência de insegurança alimentar entre 2004-2013, especialmente, quanto à insegurança alimentar moderada e grave que passou de 17% (IC95%: 15,7-18,4) em 2004 para 7,9% (IC95%: 7,2-8,7) em 2013. Por outro lado, apesar das importantes reduções na prevalência de insegurança alimentar moderada e grave, observou-se que, independentemente do nível de determinação, os estratos populacionais com menor prevalência em 2004 apresentaram redução relativa de maior magnitude. Quanto aos fatores associados à insegurança alimentar moderada e grave, permaneceram os mesmos nos dez anos cobertos pela PNAD, sendo eles: as macrorregiões Norte/Nordeste, área urbana (na presença de saneamento inadequado), densidade domiciliar > 2 pessoas/dormitório, possuir ≤ 4 bens de consumo e a pessoa de referência do domicílio ser do sexo feminino, ter idade < 60 anos, a raça/etnia ser diferente de branca, ter escolaridade ≤ 4 anos e estar desempregada. Entre 2004-2013, a prevalência de domicílios brasileiros em situação de insegurança alimentar moderada e grave caiu pela metade; contudo, dentro da perspectiva da equidade, destaca-se que os avanços ocorreram de modo desigual, sendo menores nos estratos de maior vulnerabilidade social, econômica e demográfica.

El objetivo de este estudio fue analizar la tendencia y factores asociados a la inseguridad alimentaria en Brasil durante los años 2004, 2009 y 2013, utilizando microdatos de la Encuesta Nacional por Muestra de Domicilios (PNAD). La inseguridad alimentaria se evaluó mediante la Escala Brasileña de Inseguridad Alimentaria. Las variables independientes se seleccionaron a partir de modelo conceptual de determinación de la inseguridad alimentaria, siendo utilizado también para la elaboración de los modelos lineales generalizados múltiples. Los resultados describen la tendencia de reducción en la prevalencia de inseguridad alimentaria entre 2004-2013, especialmente, en cuanto a la inseguridad alimentaria moderada y grave, que pasó de un 17% (IC95%: 15,7-18,4) en 2004, hasta un 7,9% (IC95%: 7,2-8,7) en 2013. Por otro lado, a pesar de las importantes reducciones en la prevalencia de inseguridad alimentaria moderada y grave, se observó que, independientemente del nivel de determinación, los estratos poblacionales con menor prevalencia en 2004 presentaron una reducción relativa de mayor magnitud. En cuanto a los factores asociados a la inseguridad alimentaria moderada y grave, permanecieron los mismos durante los diez años cubiertos por la PNAD, siendo estos: las macrorregiones Norte/Nordeste, área urbana (con presencia de saneamiento inadecuado), densidad domiciliaria > 2 personas/dormitorio, poseer ≤ 4 bienes de consumo y que la persona de referencia del domicilio sea del sexo femenino, tener una edad < 60 años, de raza/etnia diferente a blanca, tener escolaridad ≤ 4 años y estar desempleada. Entre 2004-2013, la prevalencia de domicilios brasileños en situación de inseguridad alimentaria moderada y grave disminuyó a la mitad; no obstante, dentro de la perspectiva de la equidad, se destaca que los avances se produjeron de modo desigual, siendo menores en los estratos de mayor vulnerabilidad social, económica y demográfica.

Anemia e fatores associados em mulheres de idade reprodutiva de um município do Nordeste brasileiro / Anemia and associated factors in women at reproductive age in a Brazilian Northeastern municipality

RESUMO: Objetivo: Determinar a prevalência de anemia e seus fatores associados em mulheres em idade reprodutiva do município de Vitória de Santo Antão, Pernambuco. Métodos: Estudo transversal com amostra representativa de 322 mulheres, com idade entre 15 e 49 anos. A anemia foi diagnosticada pela dosagem de hemoglobina abaixo de 12 g/dL. As variáveis independentes foram: condições demográficas e socioeconômicas (idade, raça/cor, escolaridade, alfabetização, presença de crianças com idade < 5 anos no domicílio, regime de ocupação de residência, posse de bens e renda familiar per capita), condições de moradia (número de pessoas no domicílio, número de dormitórios e condições de saneamento básico), segurança alimentar e nutricional e condições de saúde das mulheres (menarca, número de gestações, passado de abortos, índice de massa corporal e ferritina). A regressão de Poisson foi utilizada para a análise dos fatores associados à anemia. Resultados: A média de idade das mulheres investigadas foi de 31,1 ± 6,61 anos (desvio padrão - DP) e a prevalência de anemia, de 18,6% (IC95% 14,7 - 23,3). Após análise multivariável, a anemia apresentou associação significativa com idade ≤ 19 anos e com baixo índice de ferritina. Conclusão: A prevalência de anemia encontrada se caracteriza como um problema de saúde pública de grau leve e segue a tendência mundial de países em desenvolvimento.

ABSTRACT: Objective: To identify the prevalence of iron-deficiency anemia and associated factors in women at reproductive age in the city of Vitoria de Santo Antão, Pernambuco. Methods: Cross-sectional study with a representative sample of 322 women aged between 15 and 49 years. Anemia was diagnosed by measuring hemoglobin below 12 g/dL. The independent variables were: demographic and socioeconomic conditions (age, race/color, education, literacy, presence of children aged < 5 years in the household, regime of residence occupation, ownership of assets, and per capita income), housing conditions (number of people in the household, number of bedrooms, and basic sanitation), food and nutrition security and women's health conditions (menarche, number of pregnancies, history of miscarriages, body mass index, and ferritin). Poisson regression was used to analyze the factors associated with anemia. Results: The mean age of women surveyed was 31.1 ± 6.61 years old (standard deviation - SD) and the prevalence of anemia was 18.6% (95%CI 14.7 - 23.3). After multivariate analysis, anemia was significantly associated with age ≤ 19 years and low ferritin. Conclusion: The prevalence of anemia is characterized as a mild public health problem and follows the world trend in developing countries.

Tendência e fatores associados à insegurança alimentar no Brasil: Pesquisa Nacional por Amostra de Domicílios 2004, 2009 e 2013 / Trends and factors associated with food insecurity in Brazil: the National Household Sample Survey, 2004, 2009, and 2013 / Tendencia y factores asociados a la inseguridad alimentaria en Brasil: Encuesta Nacional por Muestra de Domicilios 2004, 2009 y 2013

O objetivo deste estudo foi analisar a tendência e fatores associados à insegurança alimentar no Brasil nos anos de 2004, 2009 e 2013, utilizando microdados da Pesquisa Nacional por Amostra de Domicílios (PNAD). A insegurança alimentar foi avaliada por meio da Escala Brasileira de Insegurança Alimentar. As variáveis independentes foram selecionadas a partir de modelo conceitual de determinação da insegurança alimentar, sendo esse também utilizado para a elaboração dos modelos lineares generalizados múltiplos. Os resultados descrevem tendência de redução na prevalência de insegurança alimentar entre 2004-2013, especialmente, quanto à insegurança alimentar moderada e grave que passou de 17% (IC95%: 15,7-18,4) em 2004 para 7,9% (IC95%: 7,2-8,7) em 2013. Por outro lado, apesar das importantes reduções na prevalência de insegurança alimentar moderada e grave, observou-se que, independentemente do nível de determinação, os estratos populacionais com menor prevalência em 2004 apresentaram redução relativa de maior magnitude. Quanto aos fatores associados à insegurança alimentar moderada e grave, permaneceram os mesmos nos dez anos cobertos pela PNAD, sendo eles: as macrorregiões Norte/Nordeste, área urbana (na presença de saneamento inadequado), densidade domiciliar > 2 pessoas/dormitório, possuir ≤ 4 bens de consumo e a pessoa de referência do domicílio ser do sexo feminino, ter idade < 60 anos, a raça/etnia ser diferente de branca, ter escolaridade ≤ 4 anos e estar desempregada. Entre 2004-2013, a prevalência de domicílios brasileiros em situação de insegurança alimentar moderada e grave caiu pela metade; contudo, dentro da perspectiva da equidade, destaca-se que os avanços ocorreram de modo desigual, sendo menores nos estratos de maior vulnerabilidade social, econômica e demográfica.

The aim of this study was to analyze trends and factors associated with food insecurity in Brazil in 2004, 2009, and 2013, using microdata from the National Household Sample Survey (PNAD). Food insecurity was assessed using the Brazilian Food Insecurity Scale. Independent variables were selected from a conceptual model of determination of food insecurity, which was also used in the elaboration of multiple generalized linear models. The results show a downward trend in food insecurity prevalence from 2004 to 2013, especially for moderate and severe food insecurity, from 17% in 2004 (95%CI: 15.7-18.4) to 7.9% in 2013 (95%CI: 7.2-8.7). Despite important decreases in the prevalence of moderate and severe food insecurity, regardless of the level of determination, the population strata with the lowest prevalence in 2004 showed the largest relative reduction. As for factors associated with moderate and severe food insecurity, they remained the same in the ten years covered by the PNAD survey, namely: the North and Northeast regions, urban areas with inadequate sanitation, household density > 2 persons per bedroom, ≤ 4 household durable consumer goods, and households headed by females, individuals < 60 years, and non-whites, ≤ 4 years of schooling, and being unemployed. From 2004 to 2013, the prevalence of Brazilian households with moderate and severe food insecurity dropped by half, but from the perspective of equity the advances occurred unequally and were lower in strata with greater social, economic, and demographic vulnerability.

El objetivo de este estudio fue analizar la tendencia y factores asociados a la inseguridad alimentaria en Brasil durante los años 2004, 2009 y 2013, utilizando microdatos de la Encuesta Nacional por Muestra de Domicilios (PNAD). La inseguridad alimentaria se evaluó mediante la Escala Brasileña de Inseguridad Alimentaria. Las variables independientes se seleccionaron a partir de modelo conceptual de determinación de la inseguridad alimentaria, siendo utilizado también para la elaboración de los modelos lineales generalizados múltiples. Los resultados describen la tendencia de reducción en la prevalencia de inseguridad alimentaria entre 2004-2013, especialmente, en cuanto a la inseguridad alimentaria moderada y grave, que pasó de un 17% (IC95%: 15,7-18,4) en 2004, hasta un 7,9% (IC95%: 7,2-8,7) en 2013. Por otro lado, a pesar de las importantes reducciones en la prevalencia de inseguridad alimentaria moderada y grave, se observó que, independientemente del nivel de determinación, los estratos poblacionales con menor prevalencia en 2004 presentaron una reducción relativa de mayor magnitud. En cuanto a los factores asociados a la inseguridad alimentaria moderada y grave, permanecieron los mismos durante los diez años cubiertos por la PNAD, siendo estos: las macrorregiones Norte/Nordeste, área urbana (con presencia de saneamiento inadecuado), densidad domiciliaria > 2 personas/dormitorio, poseer ≤ 4 bienes de consumo y que la persona de referencia del domicilio sea del sexo femenino, tener una edad < 60 años, de raza/etnia diferente a blanca, tener escolaridad ≤ 4 años y estar desempleada. Entre 2004-2013, la prevalencia de domicilios brasileños en situación de inseguridad alimentaria moderada y grave disminuyó a la mitad; no obstante, dentro de la perspectiva de la equidad, se destaca que los avances se produjeron de modo desigual, siendo menores en los estratos de mayor vulnerabilidad social, económica y demográfica.